FOOD-WEB STRUCTURE IN BIG CYPRESS SWAMP WETLANDS BASED ON STABLE-ISOTOPE RESULTS
William F.
1 USGS
1
Loftus
and David P.J.
2,3
Green
– FISC, Everglades NP, Homestead, FL, USA; 2 Audubon of Florida, Tavernier Science Center, Tavernier, FL, USA; 3 FL Gulf Coast University, Ft. Myers, FL, USA
INTRODUCTION
40.00
30.00
HQ
V
I
F
Py
A
M
40.00
30.00
Sep
05
Dec
05
Mar
06
Jun
06
Sep
06
Dec
06
Mar
07
Jun
07
Date
120
L28
100
80
60
40
20
0
-20
-40
-60
-80
-100
Jun
05
Sep
05
Dec
05
Mar
06
Jun
06
Sep
06
Dec
06
Mar
07
Jun
07
Date
100
RP
Gambusia holbrooki
Eastern mosquitofish
56.7
84.5
0
5
10 Km
40
Common Name
Grass shrimp
%
RA
%I
46.2
62
Everglades crayfish
12.4
62.7
Heterandria formosa
Least killifish
10.2
43.7
Procambarus fallax
Slough crayfish
5.6
54.9
0.00
Snail
-2.00
-4.00
-40.00
Jun
05
Sep
05
Dec
05
Mar
06
Gyrinus spp.
Whirligig water beetle
4.4
13.4
Anisoptera
Dragonfly
3.6
63.4
Coleoptera
Aquatic beetle
2.4
20.4
Coenagrionidae
Damselfly
0.7
14.8
Lucania goodei
Bluefin killifish
2.6
46.5
Lepomis gulosus
Warmouth
2.1
30.3
Labidesthes sicculus
Brook silverside
1.6
15.5
Planorbella spp.
Planorbid snail
0.7
21.8
Jun
06
Sep
06
Dec
06
Mar
07
Jun
07
Ameiurus natalis
Yellow bullhead
1.3
4.2
Corixidae
Water boatman
0.4
9.2
Poecilia latipinna
Sailfin molly
0.8
11.3
Pelocoris femoratus
Alligator flea
0.4
16.2
Dytiscidae
Predaceous diving
beetle
0.3
5.6
Lethocerus spp.
Toe biter
0.3
11.3
Physella spp.
Physid snail
0.3
8.5
Belostoma spp.
Giant water bug
0.1
7.7
Date
Ea. mosquitofish
(Gambusia holbrooki)
Sailfin molly
(Poecilia latipinna)
Least killifish
(Heterandria formosa)
Cichlasoma bimaculatum
Black acara
0.7
22.5
Elassoma evergladei
Everglades pygmy sunfish
0.6
16.9
Fundulus chrysotus
Golden topminnow
0.4
15.5
Clarias batrachus
Walking catfish
0.3
1.4
Ephemeroptera
Mayfly
0.01
1.4
Hoplosternum littorale
Brown hoplo catfish
0.2
7.7
Ranatra spp.
Water scorpion
0.01
3.5
Enneacanthus gloriosus
Bluespotted sunfish
0.2
9.2
Cichlasoma urophthalmus
Mayan cichlid
0.2
8.5
Notemigonus crysoleucas
Golden shiner
0.1
4.2
Tilapia mariae
Spotted tilapia
0.1
4.2
Lepomis punctatus
Spotted sunfish
0.1
2.1
Lepisosteus platyrhincus
Florida gar
0.01
1.4
Belonesox belizanus
Pike killifish
0.01
0.7
Shrimp
-30.00
-25.00
-20.00
-15.00
-10.00
-5.00
Table 1. Relative abundance (%RA) and % Incidence in samples of a) common fishes and b)
common invertebrates collected at the study sites (Liston et al., 2008). All taxa were
analyzed for stable-isotope values.
• Grass Shrimp
• Amphipods
-28
-23
-18
-13
0.00
Fish
Fig. 5. Comparison bi-plots for taxa with all data combined
from the Everglades (Loftus 2000) and BCNP (this study).
2005-06
CONCLUSIONS
Seasonal Comparisons: As animals move
δ15N
AMPHIPOD
8
FISH
CRAYFISH
FISH
7
INSECT
6
SNAIL
Algae
SHRIMP
SNAIL
5
PERIPHYTON
4
UTRICULARIA
DETRITUS
4
3
3
2
2
1
1
0
-33.50
-31.50
-29.50
-27.50
-25.50
-23.50
0
-33.50
-21.50
UTRICULARIA
-31.50
-29.50
-27.50
δ13C
Amphipod
-25.50
-23.50
-21.50
δ13C
Insect
8
CRAYFISH
FISH
7
INSECT
6
SHRIMP
δ15N
V - Moss
ALGAE
DETRITUS
4
UTRICULARIA
3
2
1
0
-33.50
We will explore the dataset with analytical tools such as the IsoSource (Phillips and
Gregg 2003) and circular statistical models (Schmidt et al. 2007) to examine mixing of
primary producers in the cypress. We will investigate effects of lipids on values, and will
also compare data quantitatively to food webs from other systems.
APPLICATIONS
SNAIL
5
V - Algae
FUTURE WORK
Hydrological restoration should restore aquatic food webs supporting populations
of higher vertebrates. Our data can help test that premise by defining spatio-temporal
patterns in cypress-wetland food webs as a baseline for post-restoration comparisons.
Fish
Shrimp
Isotope values varied temporally and spatially, and were similar to those from the
Everglades. Cypress dome δ13C values tended to be more depleted compared with other
south Florida systems. Both detritus and algae were food base end-members. Snails,
crayfishes, and amphipods were major 1o consumers, while fishes, shrimp, and most insects
were mainly carnivorous. As prairies dried in fall, animals moved into cypress domes but
food-web plots showed little evidence for movement. Piscivorous Florida gar had the
highest δ15N values and highest relative trophic positions. Non-native fishes functioned at
similar trophic levels to native species and used a similar range of primary producers.
RP – Wet, Dome
CRAYFISH
5
2006-07
• Gambusia / Rivulus
Vascular Pl.
-35.00
del 13C
-31.50
-29.50
-27.50
-25.50
-23.50
-21.50
RP – Transition, Dome
Figure 7. Values in wet-season marshes are enriched
compared to domes. We see little evidence of enrichment
signal in the transition-season domes, 1.5 months later, that
we would expect as animals enter the domes for dry-season
shelter. We will continue to examine these data.
-20
-40
Insect
6.00
δ15N values from Everglades animals appear enriched compared to BCNP values. Myriad factors influence δ15N values. For example,
enrichment may indicate Everglades organisms function at higher trophic levels than conspecifics in BCNP, or it may imply differences in source,
fractionation, or assimilation processes in primary producers at the base of the food web.
δ15N
• Gar / Mangrove snapper
Distinct food webs are evident in south Florida studies; data may serve to
relate spatial data to key indicator species. For example, Roseate Spoonbills
(pink circle; mean + 1 SE; N=8) receive energetic inputs from Taylor Slough
mangroves, near nesting locations and flight paths to foraging grounds
(Lorenz data).
Crayfish
8.00
10.00
Spatial/
Temporal
Comparisons
By Year
Species used in comparison:
δ13C (‰)
Amphipod
10.00
20
0
Big Cypress
12.00
Python
Fig. 4. Bi-plots of δ13C and δ15N with C:N ratios for
groups of biota. V=Vascular Plants, I=Insects,
F=Fish, Py=Python, A=Algae, and M=Moss.
Spoonbill chicks
6
Algae
2.00
-20.00
SR Mangrove Zone
14.00
20.00
-25.00
8
-33
Vascular P.
Procambarus alleni
29.6
13C
Moss
-30.00
TS Mangrove Zone
0
0.0
AMPHIPOD
40.8
2.8
-5.0
INSECT
20.8
Dollar sunfish
-10.0
6
Flagfish
Lepomis marginatus
-15.0
7
Jordanella floridae
80
60
Palaemonetes paludosus
-20.0
AMPHIPOD
Figure 2. Maps of study locations
Scientific Name
-25.0
Hydrologic restoration will affect aquatic plant communities at the food-web base.
We hypothesize that changes at the base will affect key invertebrate groups, with
consequences resonating through the web. Stable-isotope analyses will permit tracing of
food-web changes at both local and landscape scales. Our data will complement other
south Florida food-web studies to provide a more comprehensive understanding of the
entire ecosystem.
LITERATURE CITED
Liston, S. et al. 2008. Development and testing of protocols for sampling fishes in forested wetlands in southern Florida. Draft report to
USACOE and USGS-PES.
10.00
Non-native
9.00
Oscar
Pike
8.00
Fishes
del 15N
Jun
05
%I
-30.0
4.00
-35.00
10
• Snails
-4.0
-35.0
8
b) Invertebrates
% RA
10.00
del
60.00
20
-20
8.00
Freshwater ‘Glades
2
RP – Wet, Marsh
BI
Estimated water depth (cm)
6.00
12
Vascular Pl.
US 41
80
Estimated water depth (cm)
4.00
Big Cypress
4
from marsh to dome from wet to transition (Liston
Et al, 2008), do isotope values reflect those movements?
100
Common Name
2.0
Shrimp
Raccoon
Point
5
Scientific Name
Insect
I-75
N
Figure 3. BCNP water
depths showing strong
seasonal patterns at the
three study sites: Bear
Island (BI), L-28, and
Raccoon Point (RP).
4.0
BIG CYPRESS
NATIONAL
PRESERVE
Samples were field-frozen, then measured in the lab prior to drying
appropriate tissues at 50 C. Plant material was acid-treated to remove carbonate.
Dried tissue was pulverized, weighed, and prepared for analysis by mass
spectrometer at FIU (see Williams & Trexler, 2006).
0
Fish
L-28
Bear
Island
Estimated water depth (cm)
2.00
del 15 N
C:N Ratio
Primary producer, invertebrate, and fish samples of common species (Table 1) were taken
three times per year: wet season, transitional season, and dry season. Sampling was at three sites
within BCNP cypress habitats: L-28, Bear Island (BI), and Raccoon Point (RP) (Figure 2). Site
descriptions, sampling techniques, and physico-chemical data were reported in Liston et al.
(2008) (Fig. 3). Three to five individuals or sub-samples were collected for each taxon from each
sub-habitat (marsh & dome).
a) Fishes
6.0
del 13 C
METHODS AND STUDY LOCATIONS
40
Crayfish
-2.0
General
0.00
-40.00
60
Amphipod
8.0
0.0
50.00
Error bars not shown on figures for clarity of presentation.
Algae
10.0
10.00
Figure 1. Cypress Forest.
Everglades
12.0
20.00
0.00
0.00
14
δ15N (‰)
50.00
14.0
Del 15N
60.00
Patterns
Basic questions included:
What are the major primary producers?
How long are food chains?
Is there spatio-temporal variability in the web?
What roles do non-native fishes play?
V
I
F
Py
M
A
del 15N
Food webs provide maps of their biotic constituents, their roles, and interactions. We used
stable-isotope analysis to trace the movement of carbon (δ13C) from primary producers to fishes,
and to determine the trophic positions of animals by using nitrogen (δ15N).
We analyzed 777 individual fishes of 24 taxa; 7 were non-native. We also analyzed 751 invertebrate samples of
24 taxa and 733 vegetation samples of 24 taxa. Fish species names are in Table 1. Stoichiometric data for C:N showed that
as δ15N values increased, C:N ratios decreased (Fig. 4). Most vegetation taxa had δ13C values within + 2 0/00 of -30 0/00, except
algae and moss. Values from BCNP biota were remarkably similar to those from Everglades marshes (Fig. 5).
C:N Ratio
Southern Florida wetlands have been modified by drainage. A major restoration plan,
CERP, is attempting to reestablish natural hydrology. CERP success implies aquatic food webs
will be restored to support important predator populations. That premise is difficult to measure
without data on those webs. From 2005-07, we collected biotic samples to map pathways of
energy flow and trophic status of biota in freshwater wetlands of Big Cypress National Preserve
(BCNP) (Fig. 1 ). Food webs in cypress wetlands have been relatively unstudied throughout
their range.
CROSS-LANDSCAPE COMPARISONS
RESULTS:
Acara
Mayan
7.00
Clarias
Hoplo
6.00
SpotTilap
5.00
δ13C
Fig. 6. Stable-isotope values for taxa consistent among locations & seasons. The three
locations had very similar isotope values, though fish/shrimp seemed more enriched at L-28.
Most fish and insects were omnivores/carnivores. The web seems to have but three levels:
producers, 1o consumers, and 2o consumers, with both algal and detrital bases important. By the
dry season, domes had few species surviving. L-28 and RP were sites with contiguous marshes.
Native
4.00
-34.00 -33.00 -32.00 -31.00 -30.00 -29.00 -28.00 -27.00
Loftus, W.F. 2000. Mercury bioaccumulation through an Everglades aquatic food web. Ph.D. Dissert., Florida International University,
Miami.
Phillips, D.L., and J.W. Gregg. 2003. Source partitioning with stable isotopes: coping with too many sources. Oecologia 136:261-269.
Post et al. 2007. Getting to the fat of the matter: models, methods, and assumptions for dealing with lipids in stable isotope analysis.
Oecologia 152: 179-189.
Schmidt, S.N. et al. 2007. Quantitative approaches to the analysis of stable isotope food web data. Ecology 88:2793-2802.
Williams, A.J. and J.C. Trexler. 2006. A preliminary analysis of the correlation of food-web characteristics with hydrology and nutrient
gradients in the southern Everglades. Hydrobiologia 569:494-503.
del 13C
ACKNOWLEDGMENTS
Fig. 8. Introduced fish values overlapped those of native
species. Piscivorous Pike killifish had highest trophic
position, as expected.
Funding provided by USGS South Florida PES Program, administered by G.R. Best, and by Army Corps of Engineers CERP-MAP.
Special thanks to S. Liston for project assistance, to B. Dunker, N. Katin and N. Silverman for field assistance, and J. Lorenz, J. Trexler,
and C. McIvor for help in the study. We also thank K. Dunlop, L. McCarthy, W. Anderson, M. Kershaw, and C. Rebenack for sample
prep and analysis at the FIU-SERC mass-spectrometer facility.